Age-specific FSH levels as a tool for appropriate patient counselling in assisted reproduction

Transcription

1 Human Reproduction Vol.20, No.9 pp , 2005 Advance Access publication May 19, 2005 doi: /humrep/dei076 Age-specific FSH levels as a tool for appropriate patient counselling in assisted reproduction Andrea Weghofer 1,3, Markus Margreiter 1, Yvonne Fauster 2, Theresa Schaetz 2, Agnes Brandstetter 2, Doris Boehm 2 and Wilfried Feichtinger 1 1 Wunschbabyzentrum-Institute for Sterility Treatment, A-1130 Vienna and 2 Medical University Vienna, A-1090 Vienna, Austria 3 To whom correspondence should be addressed. aweghofer@gmx.at BACKGROUND: The purpose of this study was to assess whether, even within a normal FSH range (#10 mu/ml), age-specific FSH levels are predictive of ovarian reserve. METHODS: Between January 1998 and December 2001, 535 women, undergoing controlled ovarian stimulation with 225 IU of recombinant (rec) FSH and 75 IU of reclh, were included in this retrospective cohort study. Criteria for enrolment were: age years, basal FSH (b-fsh) #10 mu/ml and basal LH #12 mu/ml. Patients were assigned to three age groups (group I: years; group II: years; and group III: years). Each age group was divided into quartiles according to b-fsh levels, comparing the lowest and highest b-fsh quartiles for basal hormonal patterns and outcome-related parameters. RESULTS: At ages years, women in the lowest FSH quartiles demonstrated significantly increased numbers of oocytes at retrieval (group I: low b-fsh quartile versus high b-fsh quartile , P<0.02; group II: versus , P<0.047), whereas no difference with regard to oocyte yield was observed in patients above age 35 (group III: low b-fsh quartile versus high b-fsh quartile ). No statistical correlation was found between FSH quartiles and clinical pregnancy rates or miscarriage. CONCLUSIONS: In young women, age-specific high b-fsh levels, even within normal ranges, are associated with significantly reduced numbers of oocytes retrieved. B-FSH concentrations should, therefore, be interpreted in an age-specific manner to allow for appropriate patient counselling in IVF. Key words: age/fsh/ivf/ovarian reserve/poor responder Introduction The success of IVF is critically dependent on the appropriate evaluation of ovarian function to optimize controlled ovarian stimulation (COS). While follicular depletion and impaired oocyte quality, frequently also called diminished ovarian reserve (Kligman and Rosenwaks, 2001; Bancsi et al., 2002; Chuang et al., 2003), are generally acknowledged as the underlying mechanisms for the decline in maternal fecundity, the way to assess the status of ovarian reserve correctly has remained controversial (Fanchin et al., 1994; Lambalk, 2003). The association between advanced female age and reproductive senescence is well established (Oosterhuis et al., 2002; Jansen, 2003). Nevertheless, in women of the same age, individual variations exist with regard to their ovarian responses (Pellicer et al., 1998; Perez et al., 2000). Chronological female age is universally seen as indicative of qualitative alterations of the remaining follicle pool (Popovic-Todorovic et al., 2003; Toner, 2003; Van Rooij et al., 2003,2004a), but cannot serve as the sole marker of ovarian status (Kligman and Rosenwaks, 2001). A variety of tests, such as the antral follicle count, the clomiphene challenge test and measurements of basal inhibin B (Mukherjee et al., 1996; Creus et al., 2000; Fasouliotis et al., 2000; Meo et al., 2002), have been introduced as predictive of ovarian reserve; however, cycle day 2 or 3 FSH, i.e. basal (b-)fsh, is still the most widely used parameter in the assessment of ovarian function (Tanbo et al., 1992; Letur-Konirsch and Guis, 1996; Bancsi et al., 2003; Smits et al., 2003). Recently, some controversy has arisen about the clinical utility of b-fsh measurements (Jurema et al., 2003; Scott, 2004; Toner, 2004; Wolff and Taylor, 2004). Abdalla and Thum (2004), as well as Van Rooij et al. (2004b), depicted reduced, yet still clinically sound pregnancy and live birth rates in young women with elevated b-fsh levels of up to 20 mu/ml. They, therefore, concluded that the predictive value of b-fsh concentrations should be restricted to the counselling of patients on the probability of achieving pregnancy, but should not be used to exclude them from fertility treatment. Scott (2004) goes even further, denying any predictive value to b-fsh concentrations within normal ranges. In contrast, Jurema et al. (2003) describe substantially reduced cut off-levels for b-fsh values (, 6 IU/l) to predict ovarian response to stimulation accurately, but agree that they do not predict pregnancy potential. They also argue, 2448 q The Author Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please journals.permissions@oupjournals.org

2 Age-specific FSH as a predictor of ovarian reserve however, that b-fsh cut off-values of up to 10 IU/l will include virtually all pregnancies in a cohort with normal prognosis. Though contradictory opinions exist regarding the appropriate cut-off level for b-fsh values in the evaluation of ovarian reserve, so far there has been unanimity about interpreting b-fsh levels in absolute (i.e. fixed) terms. Since a majority of hormonal test results is subject to age-related alterations (El-Touky et al., 2002) and since ovarian reserve is known to decline in an age-dependent manner (Bancsi et al., 2003), age-specific interpretations of b-fsh values might result in a more accurate assessment of ovarian function. The study presented here, therefore, was initiated to evaluate the impact of age-specific b-fsh concentrations on ovarian response in young women with normal b-fsh levels (#10 mu/ml). Patients and methods Study design and patients Five-hundred and thirty-five regular cycling women, aged years, undergoing their first cycle of COS between January 1998 and December 2001, were included in this retrospective cohort study. Only cycles using a combination of clomiphene citrate and gonadotrophins for ovarian stimulation were eligible for enrolment. Patients suspected of having polycystic ovary syndrome were excluded from the analysis. Cut-off values for b-fsh and LH levels, measured on cycle days 2 or 3, immediately before the commencement of oral contraception, were considered as follows: b-fsh # 10 mu/ml and b-lh #12 mu/ml. Based on age, patients were assigned to the following three groups: group I: age years; group II: age years; and group III: age years. Each age group, in turn, was divided into quartiles according to b-fsh levels. Within each age group, the lowest and highest b-fsh quartiles were compared for the following parameters: female chronological age, treatment indications, b-lh and b-estradiol (E 2 ) levels, smoking status, number of oocytes retrieved, day of embryo transfer, number of embryos transferred, miscarriage and clinical pregnancy rates. A maximum of four oocytes at oocyte retrieval was considered poor ovarian response. To assess the impact of chronological female age, b-fsh level and smoking on oocyte yield within and between age groups, both a univariate and a multivariate analysis of variance (ANOVA) were performed, including the following parameters: chronological age (group I, II and III), b-fsh level (low and high b-fsh quartiles) and smoking. Furthermore, the interaction for age low b-fsh and for age high b-fsh was calculated. Ovarian stimulation and fertilization After pre-treatment with oral contraceptives (18 28 days), COS was initiated with clomiphene citrate (daily dose of 100 mg per os on cycle days 1 5) and a combination of 225 IU of recombinant (rec) FSH and 75 IU of reclh on alternate days, i.e. on cycle day 1, 3, 5 and 7 (Weigert et al., 2002). To allow for easy handling and to warrant a fixed ratio of gonadotrophins, no dosage adjustments were performed. In the case of inadequate follicular response, i.e. leading follicle diameter,18 mm after 7 days of stimulation, an additional dosage of 225 IU of recfsh and 75 IU of reclh was given on cycle day 9. Cycle monitoring, oocyte retrieval and embryo transfer were performed in a routine fashion, as previously reported (Margreiter et al., 2003). A clinical pregnancy was defined as the presence of fetal cardiac activity beyond 8 weeks of gestation. Institutional review board (IRB) approval Since the present study was based on a retrospective data analysis, no IRB approval was required. Statistical analysis The present data analysis was performed with the Statistics Package for Social Sciences (SPSS, Chicago, IL). Differences in baseline characteristics and outcome measures within and between age groups and b-fsh subsets of patients were investigated by ANOVA. Furthermore, the interaction between b-fsh (two groups: low and high b-fsh quartiles) and age (three groups: groups I III) was assessed using ANOVA. Descriptive results are presented as means with SD. s, 0.05 were considered statistically significant. Results Indications for fertility treatment and patients characteristics are presented in Tables I and II. The mean age of the women eligible for enrolment in group I was 27.5 ^ 1.2 years in the lowest b-fsh quartile and 27.8 ^ 1.2 years in the highest b-fsh quartile. In patients allocated to group II, the mean age was 32.6 ^ 1.5 in the lowest b-fsh quartile and 32.6 ^ 1.6 years in the highest b-fsh quartile, whereas women in group III were aged on average 37.5 ^ 1.4 years Table I. Indications for fertility treatment according to patients age and basal (b-)fsh levels (n = 114) a b a b a (n = 34) b Male factor 28 (70.0%) 15 (50.0%) 46 (67.6%) 33 (50.8%) 19 (55.9%) 17 (51.5%) Tubal factor 3 (7.5%) 8 (26.7%) 14 (20.6%) 21 (32.3%) 8 (23.5%) 10 (30.3%) Combined factors 2 (5.0%) 1 (3.3%) 3 (4.4%) 2 (3.1%) 0 2 (6.1%) Endometriosis 4 (10.0%) 3 (10.0%) 1 (1.5%) 2 (3.1%) 1 (2.9%) 2 (6.1%) Immunological infertility (5.9%) 0 Unexplained infertility 3 (7.5%) 3 (10.0%) 4 (5.9%) 7 (10.8%) 4 (11.8%) 2 (6.1%) a = lowest b-fsh quartile of the age group. b = highest b-fsh quartile of the age group. 2449

3 A.Weghofer et al. Table II. Patients characteristics according to female age and serum b-fsh levels (n = 114) (nthinsp; = 34) Age (years) 27.5 ^ ^ 1.2 NS 32.6 ^ ^ 1.6 NS 37.5 ^ ^ 1.5 NS b-fsh (mu/ml) 5.9 ^ ^ ^ ^ ^ ^ ^ ^ ^ 0.6 b-lh (mu/ml) 2.6 ^ ^ ^ ^ ^ ^ Estradiol (pg/ml) 19.5 ^ ^ 20.7 NS 20.4 ^ ^ 23.9 NS 20.2 ^ ^ Smoking status a 25.0% 26.7% NS 26.5% 38.4% % 39.4% NS a Statistical analysis for confounding of smoking did not reveal a significant effect on oocyte yield, either in univariate or in multivariate analysis. (lowest b-fsh quartile) and 37.5 ^ 1.5 years (highest b-fsh quartile), respectively. Age groups and FSH quartiles were similar with regard to cycle length. Within all age groups, women in the highest b-fsh quartiles showed significantly higher b-lh levels than patients in the lowest b-fsh quartiles. The correlation coefficients between b-fsh and b-lh for low versus high FSH quartiles were as follows: group I: 0.57 versus 0.14; group II: 0.42 versus 0.18; and group III: 0.32 versus At ages 36 40, women in the highest b-fsh quartile also demonstrated statistically elevated b-e 2 levels, whereas younger women showed no differences in E 2 levels between the various FSH quartiles. The impact of chronological female age and b-fsh level on oocyte yield was assessed further by univariate as well as by multivariate ANOVA. A significant influence on the number of oocytes retrieved was found for female chronological age (univariate, P, 0.001; multivariate, P = 0.039) and for age-specific high b-fsh levels (univariate, P, 0.003; multivariate, P = 0.027). To assess the impact of smoking on oocyte yield, univariate and multivariate ANOVA was performed. Smoking did not, however, have a significant impact on oocyte yield in our patient cohort (univariate analysis, P = 0.44; multivariate analysis, P = 0.56). Outcome-related parameters are summarized in Table III. Women aged years, in the lowest quartile of b-fsh levels, demonstrated significantly increased numbers of oocytes at retrieval (number of oocytes retrieved in group I, low b-fsh quartile, 8.4 ^ 3.7 versus high b-fsh quartile, 6.4 ^ 2.7, P, 0.02; group II, low b-fsh quartile, 7.5 ^ 4.0 versus high b-fsh quartile, 6.3 ^ 3.0, P, 0.047), whereas no difference with regard to oocyte yield was observed in patients above age 35 (group III, low b-fsh quartile, 5.5 ^ 3.1 versus high b-fsh quartile, 5.6 ^ 3.5, not significant). To assess further the impact of b-fsh and chronological age on oocyte yield, an interaction with main effects for age (groups I III) and b-fsh (low and high b-fsh) was performed. However, neither the interaction of low b-fsh levels and age (groups I III), nor the interaction of high b-fsh levels and age (groups I III) demonstrated significant results (low b-fsh age, P = 0.2; high b-fsh age, P = 0.6). No statistical correlation was found between clinical pregnancy rates and miscarriage to b-fsh levels Discussion A normal b-fsh level (# 10 mu/ml) and young chronological age (#35 years) are generally acknowledged as the two most promising prognostic factors, reflecting ovarian function, in women initiating fertility treatment (Jurema et al., 2003; Van Rooij et al., 2003). Such patients will, therefore, usually receive age-specific, and as mild as possible, stimulation protocols. Our study, however, clearly demonstrates that in women, aged years, b-fsh levels, even within a normal range, matter with regard to oocyte numbers. Higher age-specific FSH levels lead to fewer eggs (Figure 1). The lack of significance of the interaction between b-fsh and age supports these findings. In contrast, in women of advanced reproductive age (.35 years), the variations of b-fsh levels, as long as they are still,10.1 mu/ml, are not predictive of oocyte yield. The significant increase in b-e 2 levels in patients above age 35 with age-specific high FSH values can, though, be interpreted as reflective of diminishing ovarian reserve (Larsen et al., 2003; Tarlatzis et al., 2003; Gurbuz et al., 2004). Within our data analysis, a correlation between b-fsh and b-lh was observed in all women; patients in the highest FSH quartiles showed significantly higher LH levels than women with low age-specific FSH values. These findings might be attributable to an increasing sensitivity of the pituitary to GnRH with imminent ovarian failure, as previously reported by de Koning et al. (2000). Variations in b-fsh levels are also not predictive of clinical pregnancy and miscarriage rates. This finding should not be a surprise because, as long as patients are still capable of producing a minimal number of oocytes of acceptable quality, they will also produce adequate numbers of good quality embryos for a single embryo transfer. Consequently, high b-fsh levels, especially in young patients, should not serve as exclusion criteria from fertility treatment, but as a guidance to individual patient counselling. Large differences in pregnancy (and miscarriage) rates should become statistically visible only if the outcomes of frozen thawed embryo transfer cycles (FETs) are added in producing cumulative pregnancy rates from single retrievals. This was recently also confirmed by Gleicher et al. (personal communication) who in. 60 women with proven ovarian resistance, all below age 35, demonstrated ongoing

4 Age-specific FSH as a predictor of ovarian reserve Table III. Cycle outcomes according to female age and serum b-fsh levels (n = 144) (n = 34) No. of oocytes retrieved 8.4 ^ 3.7 (1 17) 6.4 ^ 2.7 (1 13) ^ 4.0 (1 19) 6.3 ^ 3.0 (1 15) ^ 3.1 (1 14) 5.6 ^ 3.5 (1 13) NS No. of oocyte retrievals with.4 oocytes 33 (82.5%) 24 (80%) NS 56 (82.4%) 47 (72.3%) NS 18 (52.9%) 17 (51.5%) NS No. of cycles cancelled 1 (2.5%) 2 (6.7%) NS 2 (2.9%) 1 (1.5%) NS 2 (5.9%) 3 (9.1%) NS No. of embryos transferred 2.0 ^ 0.3 (1 3) 2.0 ^ 0.5 (1 3) NS 2.3 ^ 0.6 (1 3) 2.2 ^ 0.6 (1 3) NS 2.4 ^ 0.7 (1 3) 2.1 ^ 0.6 (1 3) NS Day of embryo transfer 3.0 ^ 0.9 (2 5) 2.7 ^ 0.7 (2 4) NS. 3.1 ^ 0.9 (2 5) 2.8 ^ 0.9 (1 5) NS 3.0 ^ 0.9 (1 5) 2.6 ^ 1.0 NS Miscarriage 3 (7.5%) 6 (20%) n.s. 6 (8.8%) 3 (4.6%) NS 3 (8.8%) 4 (12.1%) NS Clinical pregnancy rate 6 (15%) 6 (20%) n.s. 18 (26.5%) 18 (27.7%) NS 6 (17.6%) 8 (24.2%) NS Figure 1. Number of oocytes retrieved according to female age and serum FSH levels. Women are allocated to groups according to chronological age; each age group is subdivided into FSH quartiles (grey, lowest FSH quartile of age group; black, highest FSH quartile of age group). Mean basal FSH (b-fsh) levels according to age groups: group I, 3.2 ^ 1.5 mu/ml in the lowest FSH quartile versus 8.8 ^ 0.7 mu/ml in the highest FSH quartile; group II, 3.3 ^ 1.4 versus 8.9 ^ 0.6 mu/ml; group III, 3.0 ^ 1.5 versus 9.0 ^ 0.6 mu/ml. pregnancy rates identical to those in matched controls with normal ovarian function. However, when cumulative pregnancy rates, indicating potential FET results, were calculated, control patients, indeed, demonstrated significantly higher pregnancy rates due to their larger oocyte numbers at the time of retrieval. Our results also coincide with those of Jurema et al. (2003) who also, at normal b-fsh levels, were unable to detect an impact on pregnancy rates in IVF cycles. Wolff and Taylor (2004) recently suggested that patients with elevated b-fsh levels should not be excluded from IVF treatment. Instead, these authors presented (in intriguing analogy to current prenatal genetic screening) that b-fsh values, among other parameters, such as chronological age and follicular phase inhibin B, should be utilized to establish an age-specific likelihood ratio for ovarian response. In conclusion, the present study demonstrates that, at young ages, b-fsh, even if within normal ranges, may contribute to the prediction of ovarian reserve, as a reflection of the number of oocytes retrieved. Consequently, b-fsh should not serve as a marker to exclude patients from fertility treatment, but should be interpreted according to patient s age and not in absolute terms, even within the generally considered normal range of # 10 mu/ml. An age-specific definition of ovarian reserve will allow physicians to stimulate patients in a more individualized fashion, which, in turn, will maximize IVF outcomes. Acknowledgements The authors thank Professor Dr Werner Brannath for his invaluable assistance with data analysis. References Abdalla H and Thum MY (2004) An elevated basal FSH reflects a quantitative rather than qualitative decline of the ovarian reserve. Hum Reprod 19,

5 A.Weghofer et al. Bancsi LF, Broekmans FJ, Eijkemans MJ, de Jong FH, Habbema JD and Te Velde ER (2002) Predictors of poor ovarian response in in vitro fertilization: a prospective study comparing basal markers of ovarian reserve. Fertil Steril 77, Bancsi LF, Broekmans FJ, Mol BW, Habbema JD and te Velde ER (2003) Performance of basal follicle-stimulating hormone in the prediction of poor ovarian response and failure to become pregnant after in vitro fertilization: a meta-analysis. Fertil Steril 79, Chuang CC, Chen CD, Chao KH, Chen SU, Ho HN and Yang YS (2003) Age is a better predictor of pregnancy potential than basal follicle-stimulating hormone levels in women undergoing in vitro fertilization. Fertil Steril 79, Creus M, Penarrubia J, Fabregues F, Vidal E, Carmona F, Casamitjana R, Vanrell JA and Balasch J (2000) Day 3 serum inhibin B and FSH and age as predictors of assisted reproduction treatment outcome. Hum Reprod 15, de Koning CH, Popp-Snijders C, Schoemaker J and Lambalk CB (2000) Elevated FSH concentrations in imminent ovarian failure are associated with higher FSH and LH pulse amplitude and response to GnRH. Hum Reprod 15, El-Toukhy T, Khalaf Y, Hart R, Taylor A and Braude P (2002) Young age does not protect against the adverse effects of reduced ovarian reserve an eight year study. Hum Reprod 17, Fanchin R, de Ziegler D, Oliviennes F, Taieb J, Dzik A and Frydman R (1994) Exogenous follicle stimulating hormone ovarian reserve test (EFORT): a simple and reliable screening test for detecting poor responders in in-vitro fertilization. Hum Reprod 9, Fasouliotis SJ, Simon A and Laufer N (2000) Evaluation and treatment of low responders in assisted reproductive technology: a challenge to meet. J Assist Reprod Genet 17, Gurbuz B, Yalti S, Ozden S and Ficicioglu C (2004) High basal estradiol level and FSH/LH ratio in unexplained recurrent pregnancy loss. Arch Gynecol Obstet 270, Jansen RP (2003) The effect of female age on the likelihood of a live birth from one in-vitro ferilization treatment. Med J Aust 178, Kligman I and Rosenwaks Z (2001) Differentiating clinical profiles: predicting good responders, poor responders, and hyperresponders. Fertil Steril 76, Jurema MW, Bracero NJ and Garcia JE (2003) Fine tuning cycle day 3 hormonal assessment of ovarian reserve improves in vitro fertilization outcome in gonadotropin-releasing hormone antagonist cycles. Fertil Steril 80, Lambalk CB (2003) Value of elevated basal follicle-stimulating hormone levels and the differential diagnosis during the diagnostic subfertility work-up. Fertil Steril 79, Larsen EC, Muller J, Schmiegelow K, Rechnitzer C and Anderson AN (2003) Reduced ovarian function in long-term survivors of radiation- and chemotherapy-treated childhood cancer. J Clin Endocrinol Metab 88, Letur-Konirsch H and Guis F (1996) Recovery of ovarian function after radiation-induced menopause. Does follicle-stimulating hormone (FSH) have a definitive prognostic value? CR Acad Sci III 319, Margreiter M, Weghofer A, Kogosowski A, Mahmoud KZ and Feichtinger W (2003) A prospective randomized multicenter study to evaluate the best day for embryo transfer. Does the outcome justify prolonged culture? J Assist Reprod Genet 20, Meo F, Ranieri DM, Khadum I and Serhal P (2002) Ovarian response and in vitro fertilization outcome in patients with reduced ovarian reserve who were stimulated with recombinant follicle-stimulating hormone or human menopausal gonadotropin. Fertil Steril 17, Mukherjee T, Copperman AB, Lapinski R, Sandler B, Bustillo M and L (1996) An elevated day three follicle-stimulating hormone: luteinizing hormone ratio (FSH:LH) in the presence of a normal day 3 FSH predicts a poor response to controlled ovarian hyperstimulation. Fertil Steril 65, Oosterhuis GJ, Vermes I, Michgelsen HW, Schoemaker J and Lambalk CB (2002) Follicle stimulating hormone measured in unextracted urine throughout the menstrual cycle correlates with age and ovarian reserve. Hum Reprod 17, Pellicer A, Ardiles G, Neuspiller F, Remohi J, Simon C and Bonilla- Musoles F (1998) Evaluation of the ovarian reserve in young low responders with normal basal levels of follicle-stimulating hormone using three-dimensional ultrasonography. Fertil Steril 70, Perez MM, Gromoll J, Behre HM, Gassner C, Nieschlag E and Simoni M (2000) Ovarian response to follicle-stimulating hormone (FSH) stimulation depends on the FSH receptor genotype. J Clin Endocrinol Metab 85, Popovic-Todorovic B, Loft A, Lindhard A, Bangsboll S, Andersson AM and Andersen AN (2003) A prospective study of predictive factors of ovarian response in standard IVF/ICSI patients treated with recombinant FSH. A suggestion for a recombinant FSH dosage normogram. Hum Reprod 18, Scott RT (2004) Diminished ovarian reserve and access to care. Fertil Steril 81, Smits G, Olatunbosun MB, Delbaere A, Pierson R, Vassart G and Costagliola S (2003) Ovarian hyperstimulation syndrome due to a mutation in the follicle-stimulating hormone receptor. N Engl J Med 349, Tanbo T, Dale PO, Lunde O, Norman N and Abyholm T (1992) Prediction of response to controlled ovarian hyperstimulation: a comparison of basal and clomiphene citrate-stimulated follicle-stimulating hormone levels. Fertil Steril 57, Tarlatzis BC, Zepiridis L, Grimbizis G and Bontis J (2003) Clinical management of low ovarian response to stimulation for IVF: a systematic review. Hum Reprod Update 9, Toner JP (2004) Modest follicle-stimulating hormone elevations in younger women: warn but don t disqualify. Fertil Steril 81, Toner JP (2003) Age = egg quality, FSH level = egg quantity. Fertil Steril 79,491. Van Rooij IA, Bancsi LF, Broekmans FJ, Looman CW, Habbema JD and te Velde ER (2003) Women older than 40 years of age and those with elevated follicle-stimulating hormone levels differ in poor response rate and embryo quality in in vitro fertilization. Fertil Steril 79, Van Rooij IA, De Jong E, Broekmans FJ, Looman CW, Habbema JD and Te Velde ER (2004a) The limited value of follicle-stimulating hormone as a test for ovarian reserve. Fertil Steril 81, Van Rooij IA, de Jong E, Broekmans FJ, Looman CW, Habbema JD and te Velde ER (2004b) High follicle-stimulating hormone levels should not necessarily lead to the exclusion of subfertile patients from treatment. Fertil Steril 81, Weigert M, Krischker U, Pohl M, Poschalko G, Kindermann C and Feichtinger W (2002) Comparison of stimulation with clomiphene citrate in combination with recombinant follicle-stimulating hormone and recombinant luteinizing hormone to stimulation with a gonadotropin-releasing hormone agonist protocol: a prospective, randomized study. Fertil Steril 78, Wolff EF and Taylor HS (2004) Value of the day 3 follicle-stimulating hormone measurement. Fertil Steril 81, Submitted on August 12, 2004; resubmitted on April 2, 2005; accepted on April 8,